Abstract
Paper presented on November 9, 1925, at a meeting of the National Academy of Sciences in Washington.
Full Text
High-Frequency Cosmic Rays1
R. A. Millikan.
As early as 1903 the English physicists Rutherford and McLennan noticed that the rate at which an electric charge leaks from an electroscope in a metal, airtight chamber decreases if the chamber is surrounded by a metal shield or box with walls one centimeter thick or more. This meant that the loss of charge in a closed electroscope was caused not by defects of insulation, but by certain rays of high penetrating power, similar to radium gamma rays. These rays can pass through more than a centimeter of metal and ionize the air inside the chamber.
In connection with this property of passing in measurable quantities through relatively thick layers of metal, the new rays were called the “penetrating radiation” of the atmosphere and at first were naturally attributed to radioactive substances on the earth. But in 1910 and 1911 it was found that the intensity of these rays does not diminish with altitude as rapidly as would follow from the hypothesis of a terrestrial origin of the unknown radiation. The first important report on this question was made by the German physicist Gockel; he ascended with a closed electroscope in a balloon to an altitude of 4 km and found that at this altitude the “penetrating radiation” had approximately the same intensity as at the surface of the earth. Meanwhile, according to the calculations of Prof. Eve, the intensity of the rays should have decreased by half already at an altitude of 75 m, if they were emitted by the earth’s surface. Two other German physicists, Hess and Kohlhörster, repeated Gockel’s measurements. Kohlhörster ascended to 9 km and found that the intensity of the radiation at first, up to about 3 km, diminishes somewhat, and then begins to increase, reaching at an altitude of 9 km a value 8 times greater than
on the surface. These observations showed that the penetrating rays come from somewhere outside and are of cosmic origin. The war interrupted investigations of this kind. After the war, in the spring of 1922, Bowen and I went to Kelly Field near San Antonio in Texas with four small, newly constructed, self-recording electroscopes. We succeeded in launching sounding balloons to an altitude of up to 15.6 km, almost twice the heights previously attained.
The instruments used were constructed for 300 cm³ of air at a pressure of 60 atmospheres and contained within them a self-recording barometer, a thermometer, and an electroscope (i.e., three separate installations, each with a moving photographic film and a driving mechanism). It is interesting that the weight of the instrument with all its parts was only 180 g.
On the basis of the above-mentioned results of Kolhörster, we expected to find in our experiments a very large value for the discharge of the electroscopes. Our instruments rose to such an altitude that \(9/10\) of the atmosphere (with respect to mass) already remained below; the attenuation of the intensity of the hypothetical rays coming from outside could here occur only through absorption in \(1/10\) of the atmosphere. Our experiments convincingly showed, in agreement with the observations of Kolhörster and Hess, that the intensity of the penetrating rays at altitude was greater than at the surface. However, at the limiting altitude we found an increase only about one quarter of the value that could have been expected from the data of the German investigators. (Two years later, on the basis of further experiments, they somewhat corrected their data, which now no longer contradict our measurements.)
The origin of the rays nevertheless remained unclear, although there were grounds for assuming their cosmic nature. Mr. Russel Otis and I believed that it was necessary to determine the degree of penetration of the rays. The rays were weaker below than above. Therefore in the summer of 1923 we set out for the summit of Pike’s Peak, taking with us 120 kg of lead and a large water tank (\(2.7 \times 2.7 \times 2.7\) m) in order to measure the absorption of the rays at altitude.
We found that the electroscopes on the mountain discharged twice as fast as in Pasadena; however, the rays were very strongly absorbed by our shields; the penetrating power of the greater part of these rays did not exceed the penetrating power of ordinary gamma rays of radium. We further found that the rate of discharge of our electroscope decreased by 10 percent owing to a snowstorm that occurred during our stay on the mountain. It clearly followed from this that, at least, the greater part of the rays with which we experimented on the mountain was of local origin; these rays could have been caused by radioactive substances carried in some unknown way into the upper layers of the atmosphere.
HIGH-FREQUENCY COSMIC RAYS
Clarifying the reason for the increase in the intensity of these soft rays (similar to gamma rays) with altitude became a task just as interesting as the question of the existence of cosmic radiation with great penetrating power. These penetrating rays could account for only a very small fraction of the increase in intensity that we observed when ascending from the level of Pasadena to Pike’s Peak. Therefore, in the summer of 1925 Mr. Harvey Cameron and I decided to carry out new experiments. The purpose of these experiments was: 1) to decide definitively the question of the existence or nonexistence of rays of great penetrating power of cosmic origin; 2) to determine the cause of the change in intensity, observed by us, of soft radiation of the gamma-ray type with altitude.
To detect penetrating rays, if such rays exist, it was necessary to find, at a very great altitude, very deep lakes formed from melted snow. Any radioactive contamination of the water, during its wandering underground, could have distorted the results that we hoped to obtain by immersing electroscopes to various depths in the lake.
For the first experiment we chose the beautiful Lake Muir, at an altitude of 3.5 km, several tens of meters deep, located precisely below Mount Whitney, the highest mountain in the U.S. Here we worked during the last 10 days of August 1925, immersing our electroscopes to various depths down to 18 m. Our experiments revealed, quite indisputably, cosmic radiation of such extraordinarily penetrating power that the fall of the electroscope charge could be traced down to a depth of 14 m. The atmosphere above the lake, in absorbing power, was equivalent to 7 m of water. Thus we discovered rays arriving at the earth from outer space with such penetrating power that they overcame a layer of \(14 + 7 = 21\) m of water, which is equivalent to 180 cm of lead. The hardness of these rays thus surpasses everything that could have been imagined.
The most penetrating X-rays of medical installations pass through no more than 1 cm of lead. The new rays surpass X-rays in their penetrating power hundreds of times.
According to modern physical views, great penetrating power corresponds to high frequency, or short wavelength. Our experiments show that there exists a range of frequencies exceeding the frequencies of X-rays by approximately as many times as X-rays exceed visible waves in their frequency. The experiments show quite definitely that these rays of enormous frequency are not homogeneous, but are distributed over a certain spectral region. The shortest waves observed by us are almost twice as short as the longest. We conclude this because the hardness
R. A. MILLIKAN
or the frequency of the rays observed at Lake Muir changed as they were filtered through ever greater and greater thicknesses of water, just as X-rays become increasingly hard when passed through successive layers of lead. Experiments with sounding balloons prove that the frequencies of these cosmic rays do not fall within the region of X-rays; otherwise we would have obtained a considerably greater rate of fall of the charge of electroscopes at those altitudes where only 0.1 atmosphere remains.
Further, we have obtained reliable proof that the new rays fly through space in all directions; this is evidenced by the constancy of the intensity of these rays both by day and by night.
All these results, obtained at Lake Muir, were confirmed with astonishing accuracy in another series of observations on another snow-fed lake, Arrowhead, 2.1 km below Lake Muir. The depth of this lake is approximately the same as at Lake Muir. The absorbing capacity of the atmosphere corresponding to the difference in height between Lakes Muir and Arrowhead is approximately equivalent to 2 m of water. Indeed, each measurement at Lake Arrowhead practically gave the same figures as at Lake Muir, but at a depth two meters lower.
One may draw certain quite probable conclusions about the origin of these penetrating high-frequency rays. The hardest rays that we have known up to now, the gamma rays of radium and thorium, arise only in changes in the nucleus of the atom. In other words, they are produced in the transformations of one atom into another, or in the creation of an atom of a new type. Therefore it is hardly possible to avoid the conclusion that the new, still more penetrating rays which we have studied originate in transformations of a similar type. However, these transformations must be many times more energetic than the processes we have known up to now. According to our new views, the energy of intra-atomic transformations must be proportional to the frequency of the light emitted in them. Therefore, apparently, it is inevitable to suppose that the changes of the nucleus corresponding to the emission of penetrating rays are accompanied by the release of energy approximately 50 times greater than in ordinary radioactive processes. These transformations of nuclei occur throughout space, and signals of them reach us in the form of penetrating rays.
The energy corresponding to the formation of helium from hydrogen is known. From it we calculated the corresponding frequency and found that it very nearly coincides with the greatest frequency of the penetrating rays found by us this summer1. The frequencies of these rays found
HIGH-FREQUENCY COSMIC RAYS
...also nearly coincide with the frequency emitted in the simple capture of an electron by a positive nucleus1). It is possible that a process of this kind takes place throughout all space. I believe that this is the most probable source of the new rays. True, the frequency has to be calculated from a formula relating the frequency of the rays to the absorption coefficient; how accurate this formula is is still uncertain. It is justified, however, in entirely independent regions of X-radiation and gamma rays.
On the basis of this formula, the shortest wavelength found by us is \(0.0004\ \text{\AA}\), i.e., approximately 50 times shorter than the hardest gamma rays known up to now. The longest wave of the penetrating radiation is \(5/3\) of the shortest, i.e. \(0.00067\ \text{\AA}\).
When these extraordinarily hard rays meet the earth, then, according to the now firmly established Compton effect, they are partly transformed into softer rays, of almost the same hardness, or wavelength, as that which we observed on Mount Pike and Mount Whitney. The reason for the greater intensity of these soft rays on mountain summits than in Pasadena is explained simply by the fact that the intensity of the primary hard rays that are transformed into soft ones at these altitudes is approximately twice as great as in Pasadena. Such, in our view, is the solution of the second problem posed by us this summer.
But how can the radiation accompanying the formation of helium from hydrogen, or the capture of an electron by a positive nucleus, be propagated throughout all space? We obtained completely identical results both at noon and at midnight; therefore the rays, apparently, come with equal intensity from all directions, independently of the position of the sun. This difficulty is not so great if one takes into account the transparency even of very large accumulations of matter for these hard rays. According to the data of Hubble, obtained recently at the Mount Wilson Observatory, some spiral nebulae are at least a million light-years distant from us, and nevertheless visible light from them reaches us. The centers of the indicated nuclear changes must be only very sparsely scattered throughout
space; this is sufficient to explain the intensity of penetrating rays with which we were dealing at Lake Muir.
Another hypothesis, besides the one indicated, could only proceed from the supposition that penetrating rays are formed in the upper layers of the atmosphere by electrons that traverse space in all directions practically at the speed of light. Such a hypothesis could explain the mysterious constancy of the earth’s negative charge; however, it encounters insurmountable difficulties: namely, it cannot quantitatively explain the changes of ionization with altitude in closed vessels.
In any case, this second hypothesis is, in its essential features, very similar to the first, and in both cases space must be conceived as filled with rays of one kind or another, flying in all directions at the speed of light. There appear to be no other alternatives besides those indicated. Such a conception may serve as a powerful stimulus to the imagination. Prof. MacMillan (Chicago) wishes to see here proof of the transformation of light into matter in space, and psychologists, perhaps, will explain telepathic phenomena of every kind by new rays.
In any case, our experiments allow us to draw the following conclusions: 1) penetrating rays exist; 2) their mass absorption coefficient is 0.18 per 1 m of water; 3) they are heterogeneous, but are distributed over a spectral region lying very far from the region of X-rays—their frequency is approximately 1000 times greater than the mean frequency of X-rays; 4) these hard rays, on encountering matter, are transformed into softer ones, whose frequency is predicted by the theory of the Compton effect; 5) penetrating rays reach the earth with the same intensity by day and by night and at all hours of the day and night; their intensity is practically the same in all directions.